Inside Toru Iwatani's Philosophy — Build a Maze, Then Hide That It Is One

Introduction

Toru Iwatani is the man who planned Pac-Man at Namco in 1980. Almost everyone knows the name; far fewer could say in one sentence what he was actually thinking about when he designed. The famous line — that he built it to bring women into arcades — tends to be where the reading stops rather than where it starts.

I am putting him on the desk now because Susan Lammers' 1986 book Programmers at Work has been announced for a 40th Anniversary Edition in July 2026 (Programmers At Work site). That book contains an interview with Iwatani at thirty-one. Which means the 1986 Iwatani, the Iwatani who stood on the GDC stage in 2011, and the Iwatani who lectured as a university professor in 2015 can all be laid on the same desk and read against each other.

Line them up and something snags. This is a man who spent a career preaching that a player should understand the objective at a glance — and in 2011 he criticised a flood of very simple games. The person who built the outer limit of simplicity, scolding simplicity. I refilled the hojicha and went after how those two live in the same head. What follows is not a review of Pac-Man; it is a study of the man.

Background — a planner inside the company that started with a coin-op horse

Iwatani was born in 1955. He joined Namco and worked as a video game planner. His first project was Gee Bee (1978), a hybrid of pinball and video game, followed by Cutie Q and Bomb Bee. Then, in 1980, Pac-Man. He stayed in planning and production roles for decades before moving to a university.

In a 2015 interview he described where he had landed: "I left the video game industry to work at a university, and to pursue the study of Game Theory for social good," and "Currently, I am teaching students Game Theory, Game Planning, and Serious Games in the Department of Game, Faculty of Arts at Tokyo Polytechnic University" (Geek Culture, 2015). He left the industry to handle play as a subject of study instead.

His account of the company's origins is telling too. In a 2016 talk he said Namco is "a company that began with VR," tracing a single line from its first product — a coin-operated electric rocking horse at five yen a play — through to its VR research (4Gamer, 2016). A man who shelves a rocking horse next to a headset is worth noting: his design thinking sorts things not by what is happening on screen but by what a person feels.

Philosophy — subtract, make it legible at a glance, make failure make sense

The core of Iwatani's philosophy comes out plainly in a lecture he gave on 28 October 2015, hosted by the Japan Online Game Association, titled "To make a game is to know the human heart." There he argued that one should think by subtraction rather than addition, and said Pac-Man was built by thinking it through "until nothing could be added and nothing could be taken away" (4Gamer, 2015).

In the same lecture he listed conditions for good design: the objective must be clear at a glance; the reason for a mistake must be clearly understandable; the difficulty curve needs landings — moments where the player gets to have the game in hand. Underneath sits an assumption about people rather than about games. Humans, he says, put fun first, and do not generally welcome things that are hard or painful. Before the design rules comes the recognition that people do not endure much.

The same argument appears in his GDC 2011 talk. Push players to the edge of giving up, then reward them: a reporter summarised the point as "players should approach giving up, then get rewarded with an easier section once they clear that hardship" (Game Developer, 2011).

Now go back to 1986. In the Lammers interview he explained the ghosts' wave attacks like this: "When a human being is constantly under attack like this, he becomes discouraged. So we developed the wave-patterned attack—attack then disperse" (Programmers at Work, 1986), going on to describe the ghosts regrouping over time as the peaks and valleys of the wave flatten out. The "wave" of 1986 and the "landing" of 2015 are the same idea in different words. Thirty years, and he has not moved on this point.

Obsessions — start from a verb, hide the maze, split the tempers

The first obsession is starting from a verb. In his 2016 talk he explained that in system design he "thinks of the verb first," and reproduced the path to Pac-Man as: for women… fashion… romance… eating! (4Gamer, 2016). At GDC 2011 he told the same story in more domestic terms: "girls love to eat desserts – my wife often eats desserts. … The verb eat, that gave me a hint to create this game" (Game Developer, 2011). Not shoot, not jump. Eat. Everything else follows from that.

The second is my favourite thing about his design. He made a maze game and then worked to keep the maze from being noticed. "I thought some people would be put off by the word maze, so I made the walls thin and blue like neon tubing, unobtrusive, on a black background so they would blend quietly into it" (Bandai Namco, "Chishin" interview). In 2016 he said again that the walls are drawn as outlines only so the game would not read as "a tedious maze game" (4Gamer, 2016). The maze is necessary; the signage announcing it is not. For anyone who makes puzzles, that is a heavy call.

The third is distributing temperaments among the pursuers. The four ghosts carry the internal names Oikake (chaser), Machibuse (ambusher), Kimagure (whimsical) and Otoboke (feigning ignorance), and each moves differently (Bandai Namco, "Chishin" interview). His 2015 lecture is more specific: red always pursues Pac-Man, pink aims 32 dots ahead of him, blue targets a point-symmetric position, orange moves at random. Make them all behave like red, he said, and the game stops being interesting (4Gamer, 2015).

Note that only one of these three is subtraction. Narrowing to a single verb is subtraction; giving four pursuers four separate minds is plainly addition. What he cuts is the amount the player must understand; what he grows is the amount the board can answer back with. Minimum vocabulary on Pac-Man's side, maximum on the ghosts'. That asymmetry shows that his "subtraction" was never about looking simple.

Failure, and how he took it — Gee Bee was too hard

Among the failures Iwatani has discussed publicly, one bears directly on his design theory: his own account of his debut game, Gee Bee. "With Gee Bee I made it too hard. Initial sales grew, but revenue dropped off sharply straight after" (Bandai Namco, "Chishin" interview). The numbers came, then they fell. He himself attributes the fall to difficulty.

Read that sentence before you read his theory of difficulty and the order reverses. "People do not welcome hard things," "build in landings," "reward them with an easier stretch once they nearly quit" — none of these arrive from an abstract view of human nature. They read as a working lesson from someone who once watched a revenue graph fall, and who has been repeating it for thirty years since.

His method of verification extends the same practicality. On internal testing he said that whether colleagues played it during their lunch break was "one of the barometers of popularity" (Bandai Namco, "Chishin" interview, part 2). Not whether they play it dutifully, but whether they reach for it on their own time. The location-test story has the same texture: seeing women playing and shrieking with delight, he judged that the concept had landed exactly as planned (part 1).

Dilemmas — the pressure to paint them all red, and the scolding of simplicity

The first dilemma he has described is getting an authored decision through an organisation. According to his 2015 lecture, a superior instructed him to make all the ghosts red. Rather than argue head-on, he surveyed forty people in the company, found unanimous support for the colourful version, and used that data to persuade management. His lesson: understand the other party's values and choose the argument that works on that party (4Gamer, 2015).

What is interesting is that in the same lecture he argues the opposite as a theory of leadership. A leader's chief job is deciding; when a team splits into plan A and plan B, the leader should decide immediately by authority, and an instruction to go compare data at length risks a quagmire — "you also need the resolve to accept that you may be wrong." The young man who moved his bosses with data and the older man who warns against ordering data comparisons stand on the same stage. Less a contradiction than an honest record of changing seats.

The second is the snag I put at the top of this piece. In 2011, as a warning to the industry, he said: "When you look at games coming out today, it's doubtful that any of us will be talking about them in ten years' time," and gave his reason: "The reason I want to emphasize this is that starting last year or so, you've had this flood of very simple games on the iPhone and social networks and so forth" (Game Developer, 2011).

Pac-Man is one four-way stick and an objective legible at a glance — a specimen of simplicity. And its author dislikes a flood of simple games. But look closely at the words he uses in that same piece: what he asks for is games that are "very detailed, yet neat and approachable," and he says developers should leave profit maximisation to management and "think more about what games mean to them." What he is scolding is not simplicity but the absence of anything behind it — that, at least, is as far as his own words will carry.

Influences — Huizinga, Caillois, Maslow, and a plush toy on a train

The theoretical grounding he acknowledges is gathered in his 2015 lecture: he says his research builds on Huizinga's account of the human as a playing being, Caillois' four categories of play, and Maslow's hierarchy of needs (4Gamer, 2015). A working planner trying to underwrite his practice from the direction of play theory and motivation theory.

The other influence is a habit rather than a theory. In the same lecture he preaches everyday observation. See someone on a train holding a plush toy, and you observe which character is in fashion, what size is preferred, what kind of person is carrying it — and you build a hypothesis. He also says an ideation meeting needs an atmosphere where no suggestion is shot down, and that ideas which made the room laugh in disbelief frequently turned into hits.

And the direct seed of Pac-Man lay inside that habit of observation. A domestic observation — his wife often eats desserts — turned into the verb eat (Game Developer, 2011). His description of the arcades of the day is observation too: "That was a playground for boys. It was dirty and smelly." He wanted them brighter and cleaner. The starting point of the design was not a work of art but a place and the people in it.

Kizuki's reading — a designer who works as an observer, not an author

What follows is my own reading. If you read Iwatani as "the man of simplicity," you will always trip over his 2011 criticism. I read him instead as a relentless observer. His design never begins from inside the work. A dirty, smelly arcade; a wife eating dessert; colleagues who do or do not reach for the cabinet at lunch; a stranger on a train holding a plush toy; the revenue graph of Gee Bee falling away. Every one an observation from outside. That is why outline walls were enough — he had already watched people flinch at the word maze. Simplicity was not his aesthetic; it was the shape his answer to an observation happened to take. Place him there and the 2011 scolding coheres: what angered him was a way of making that stopped observing on the near side of simplicity, not simplicity itself. And his insistence that Pac-Man was "made with the cooperation of a great many people, starting with those here today," refusing excess credit for himself (Bandai Namco, "Chishin" interview, part 2), reads as the consistent posture of an observer who declines to call himself the author.

In closing — where to start

If you want to understand Iwatani, play Pac-Man a few times while watching only one thing: how the four ghosts differ. There is a moment, while red is chasing you, when you notice pink has gone around ahead. The design of that noticing is the concrete form of his rule that the reason for a mistake must be clearly understandable. Read the Gee Bee story afterwards and the feel of why he built landings into the difficulty changes.

As a reading order, the three most useful stops are the 1986 Lammers interview, the 2011 GDC talk, and the 2015 JOGA lecture, in that order. Thirty years of words that have barely moved. What moves is only his seat: planner, producer, and then a university teacher who says he wants to make games that reach everyone, including people who cannot see (4Gamer, 2015).

For adjacent reading on this site, try Alexey Pajitnov (never letting a game of making order end) and Hiroyuki Imabayashi (the cargo you are tidying gets in the way of tidying). All three design almost nothing on the board and almost everything on the human side of it.

Sources

Primary sources used in this article:

・Programmers At Work official site, "Toru Iwatani, 1986 PacMan Designer" (landing page for the Iwatani interview in Susan M. Lammers, Programmers at Work, 1986; a 40th Anniversary Edition is announced for July 2026)

・Game Developer, "GDC 2011: Pac-Man Creator Iwatani Discusses Female-Friendly Design Lessons", 2 March 2011 (report on his own GDC 2011 talk)

・Game Developer, "Pac-Man Creator Iwatani's 'Cautionary Message For The Industry'", 24 June 2011

・GDC Vault, "Classic Game Postmortem: PAC-MAN", GDC 2011 (video of Iwatani's own session)

・Bandai Namco Funfare, "Bandai Namco Chishin #7: The Birth of Pac-Man, Part 1" (interview with Toru Iwatani and other developers)

・Bandai Namco Funfare, "Bandai Namco Chishin #7: The Birth of Pac-Man, Part 2" (same series)

・4Gamer, report on Toru Iwatani's JOGA seminar "To make a game is to know the human heart", 29 October 2015 (lecture given 28 October 2015)

・4Gamer, report on "Night GAME ON, third evening" with Toru Iwatani and Masanobu Endo, 31 May 2016 (talk held 27 May 2016)

・Geek Culture, "EXCLUSIVE: Interview with Toru Iwatani, creator of Pac-Man", 22 May 2015

A related interview with him, not quoted in this article:

・Journal of the Japanese Society for Artificial Intelligence, Vol. 34 No. 1, "How was Pac-Man, the origin of game AI, created?: An interview with Toru Iwatani" (interviewer: Youichiro Miyake, 2019)

Supplementary references:

・Wikipedia (Japanese), "Toru Iwatani (game creator)" (birth year, credits, affiliation)

・Wikipedia, "Tōru Iwatani" (career details)